Health-system readiness for the care of preterm, small and sick newborns in Western Kenya: a mixed-methods assessment of facilities, workforce and service delivery

Abstract

Background Preterm, small and sick newborns depend on facilities that are not only equipped but staffed, protocol-guided and able to refer, record and improve. Multidomain readiness evidence at county level in Kenya is limited. We assessed readiness in selected facilities in Western Kenya and examined health-system factors explaining the gaps. Methods Facility-level cross-sectional mixed-methods study in ten facilities (levels 3–5) in Bungoma and one neighbouring county. A structured assessment scored 59 items across 11 domains (0 = not available; 1 = available but non-functional; 2 = fully available and functional). Domain and overall readiness, tracer functionality and eight service bundles were computed from observed scores. Two provider interviews and facility free-text responses were analysed using a framework approach and integrated in a joint display. Results No item-level score was missing. Median overall readiness was 49.2% (IQR 45.8–72.7; range 29.7–94.9), with five facilities below 50%. Readiness was highest for health information systems (median 75.0%) and infrastructure (71.4%) and lowest for kangaroo mother care (20.0%), clinical guidelines (30.0%) and quality improvement (31.2%). No facility had an adequate nurse-to-newborn ratio; a preterm care guideline was fully functional in one. Two facilities met the full stabilisation bundle, one the respiratory bundle, and one met all four core bundles. Providers attributed gaps to staffing shortages, donor dependence and audit recommendations that recur unimplemented. Conclusions In this network, commodities were more available than the workforce, protocols and governance needed to use them. Priority investment should target neonatal staffing, protocol dissemination, KMC capacity and facility quality-improvement structures rather than equipment alone.